op-geth/les/peer.go

814 lines
24 KiB
Go
Raw Normal View History

2016-11-09 01:01:56 +00:00
// Copyright 2016 The go-ethereum Authors
2016-10-14 03:51:29 +00:00
// This file is part of the go-ethereum library.
//
// The go-ethereum library is free software: you can redistribute it and/or modify
// it under the terms of the GNU Lesser General Public License as published by
// the Free Software Foundation, either version 3 of the License, or
// (at your option) any later version.
//
// The go-ethereum library is distributed in the hope that it will be useful,
// but WITHOUT ANY WARRANTY; without even the implied warranty of
// MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
// GNU Lesser General Public License for more details.
//
// You should have received a copy of the GNU Lesser General Public License
// along with the go-ethereum library. If not, see <http://www.gnu.org/licenses/>.
package les
import (
"errors"
"fmt"
"math/big"
"sync"
"time"
2016-10-14 03:51:29 +00:00
"github.com/ethereum/go-ethereum/common"
les, les/flowcontrol: improved request serving and flow control (#18230) This change - implements concurrent LES request serving even for a single peer. - replaces the request cost estimation method with a cost table based on benchmarks which gives much more consistent results. Until now the allowed number of light peers was just a guess which probably contributed a lot to the fluctuating quality of available service. Everything related to request cost is implemented in a single object, the 'cost tracker'. It uses a fixed cost table with a global 'correction factor'. Benchmark code is included and can be run at any time to adapt costs to low-level implementation changes. - reimplements flowcontrol.ClientManager in a cleaner and more efficient way, with added capabilities: There is now control over bandwidth, which allows using the flow control parameters for client prioritization. Target utilization over 100 percent is now supported to model concurrent request processing. Total serving bandwidth is reduced during block processing to prevent database contention. - implements an RPC API for the LES servers allowing server operators to assign priority bandwidth to certain clients and change prioritized status even while the client is connected. The new API is meant for cases where server operators charge for LES using an off-protocol mechanism. - adds a unit test for the new client manager. - adds an end-to-end test using the network simulator that tests bandwidth control functions through the new API.
2019-02-26 11:32:48 +00:00
"github.com/ethereum/go-ethereum/common/mclock"
2016-10-14 03:51:29 +00:00
"github.com/ethereum/go-ethereum/core/types"
"github.com/ethereum/go-ethereum/eth"
"github.com/ethereum/go-ethereum/les/flowcontrol"
"github.com/ethereum/go-ethereum/light"
2016-10-14 03:51:29 +00:00
"github.com/ethereum/go-ethereum/p2p"
"github.com/ethereum/go-ethereum/rlp"
)
var (
errClosed = errors.New("peer set is closed")
errAlreadyRegistered = errors.New("peer is already registered")
errNotRegistered = errors.New("peer is not registered")
errInvalidHelpTrieReq = errors.New("invalid help trie request")
2016-10-14 03:51:29 +00:00
)
const maxResponseErrors = 50 // number of invalid responses tolerated (makes the protocol less brittle but still avoids spam)
2016-10-14 03:51:29 +00:00
les, les/flowcontrol: improved request serving and flow control (#18230) This change - implements concurrent LES request serving even for a single peer. - replaces the request cost estimation method with a cost table based on benchmarks which gives much more consistent results. Until now the allowed number of light peers was just a guess which probably contributed a lot to the fluctuating quality of available service. Everything related to request cost is implemented in a single object, the 'cost tracker'. It uses a fixed cost table with a global 'correction factor'. Benchmark code is included and can be run at any time to adapt costs to low-level implementation changes. - reimplements flowcontrol.ClientManager in a cleaner and more efficient way, with added capabilities: There is now control over bandwidth, which allows using the flow control parameters for client prioritization. Target utilization over 100 percent is now supported to model concurrent request processing. Total serving bandwidth is reduced during block processing to prevent database contention. - implements an RPC API for the LES servers allowing server operators to assign priority bandwidth to certain clients and change prioritized status even while the client is connected. The new API is meant for cases where server operators charge for LES using an off-protocol mechanism. - adds a unit test for the new client manager. - adds an end-to-end test using the network simulator that tests bandwidth control functions through the new API.
2019-02-26 11:32:48 +00:00
// capacity limitation for parameter updates
const (
allowedUpdateBytes = 100000 // initial/maximum allowed update size
allowedUpdateRate = time.Millisecond * 10 // time constant for recharging one byte of allowance
)
// if the total encoded size of a sent transaction batch is over txSizeCostLimit
// per transaction then the request cost is calculated as proportional to the
// encoded size instead of the transaction count
const txSizeCostLimit = 0x10000
const (
announceTypeNone = iota
announceTypeSimple
announceTypeSigned
)
2016-10-14 03:51:29 +00:00
type peer struct {
*p2p.Peer
rw p2p.MsgReadWriter
version int // Protocol version negotiated
network uint64 // Network ID being on
2016-10-14 03:51:29 +00:00
announceType uint64
2016-10-14 03:51:29 +00:00
id string
headInfo *announceData
lock sync.RWMutex
2016-10-14 03:51:29 +00:00
les, les/flowcontrol: improved request serving and flow control (#18230) This change - implements concurrent LES request serving even for a single peer. - replaces the request cost estimation method with a cost table based on benchmarks which gives much more consistent results. Until now the allowed number of light peers was just a guess which probably contributed a lot to the fluctuating quality of available service. Everything related to request cost is implemented in a single object, the 'cost tracker'. It uses a fixed cost table with a global 'correction factor'. Benchmark code is included and can be run at any time to adapt costs to low-level implementation changes. - reimplements flowcontrol.ClientManager in a cleaner and more efficient way, with added capabilities: There is now control over bandwidth, which allows using the flow control parameters for client prioritization. Target utilization over 100 percent is now supported to model concurrent request processing. Total serving bandwidth is reduced during block processing to prevent database contention. - implements an RPC API for the LES servers allowing server operators to assign priority bandwidth to certain clients and change prioritized status even while the client is connected. The new API is meant for cases where server operators charge for LES using an off-protocol mechanism. - adds a unit test for the new client manager. - adds an end-to-end test using the network simulator that tests bandwidth control functions through the new API.
2019-02-26 11:32:48 +00:00
sendQueue *execQueue
errCh chan error
// responseLock ensures that responses are queued in the same order as
// RequestProcessed is called
responseLock sync.Mutex
responseCount uint64
2016-10-14 03:51:29 +00:00
poolEntry *poolEntry
hasBlock func(common.Hash, uint64, bool) bool
responseErrors int
les, les/flowcontrol: improved request serving and flow control (#18230) This change - implements concurrent LES request serving even for a single peer. - replaces the request cost estimation method with a cost table based on benchmarks which gives much more consistent results. Until now the allowed number of light peers was just a guess which probably contributed a lot to the fluctuating quality of available service. Everything related to request cost is implemented in a single object, the 'cost tracker'. It uses a fixed cost table with a global 'correction factor'. Benchmark code is included and can be run at any time to adapt costs to low-level implementation changes. - reimplements flowcontrol.ClientManager in a cleaner and more efficient way, with added capabilities: There is now control over bandwidth, which allows using the flow control parameters for client prioritization. Target utilization over 100 percent is now supported to model concurrent request processing. Total serving bandwidth is reduced during block processing to prevent database contention. - implements an RPC API for the LES servers allowing server operators to assign priority bandwidth to certain clients and change prioritized status even while the client is connected. The new API is meant for cases where server operators charge for LES using an off-protocol mechanism. - adds a unit test for the new client manager. - adds an end-to-end test using the network simulator that tests bandwidth control functions through the new API.
2019-02-26 11:32:48 +00:00
updateCounter uint64
updateTime mclock.AbsTime
2016-11-17 14:54:24 +00:00
les, les/flowcontrol: improved request serving and flow control (#18230) This change - implements concurrent LES request serving even for a single peer. - replaces the request cost estimation method with a cost table based on benchmarks which gives much more consistent results. Until now the allowed number of light peers was just a guess which probably contributed a lot to the fluctuating quality of available service. Everything related to request cost is implemented in a single object, the 'cost tracker'. It uses a fixed cost table with a global 'correction factor'. Benchmark code is included and can be run at any time to adapt costs to low-level implementation changes. - reimplements flowcontrol.ClientManager in a cleaner and more efficient way, with added capabilities: There is now control over bandwidth, which allows using the flow control parameters for client prioritization. Target utilization over 100 percent is now supported to model concurrent request processing. Total serving bandwidth is reduced during block processing to prevent database contention. - implements an RPC API for the LES servers allowing server operators to assign priority bandwidth to certain clients and change prioritized status even while the client is connected. The new API is meant for cases where server operators charge for LES using an off-protocol mechanism. - adds a unit test for the new client manager. - adds an end-to-end test using the network simulator that tests bandwidth control functions through the new API.
2019-02-26 11:32:48 +00:00
fcClient *flowcontrol.ClientNode // nil if the peer is server only
fcServer *flowcontrol.ServerNode // nil if the peer is client only
fcParams flowcontrol.ServerParams
fcCosts requestCostTable
isTrusted bool
isOnlyAnnounce bool
2016-10-14 03:51:29 +00:00
}
func newPeer(version int, network uint64, isTrusted bool, p *p2p.Peer, rw p2p.MsgReadWriter) *peer {
2016-10-14 03:51:29 +00:00
return &peer{
les, les/flowcontrol: improved request serving and flow control (#18230) This change - implements concurrent LES request serving even for a single peer. - replaces the request cost estimation method with a cost table based on benchmarks which gives much more consistent results. Until now the allowed number of light peers was just a guess which probably contributed a lot to the fluctuating quality of available service. Everything related to request cost is implemented in a single object, the 'cost tracker'. It uses a fixed cost table with a global 'correction factor'. Benchmark code is included and can be run at any time to adapt costs to low-level implementation changes. - reimplements flowcontrol.ClientManager in a cleaner and more efficient way, with added capabilities: There is now control over bandwidth, which allows using the flow control parameters for client prioritization. Target utilization over 100 percent is now supported to model concurrent request processing. Total serving bandwidth is reduced during block processing to prevent database contention. - implements an RPC API for the LES servers allowing server operators to assign priority bandwidth to certain clients and change prioritized status even while the client is connected. The new API is meant for cases where server operators charge for LES using an off-protocol mechanism. - adds a unit test for the new client manager. - adds an end-to-end test using the network simulator that tests bandwidth control functions through the new API.
2019-02-26 11:32:48 +00:00
Peer: p,
rw: rw,
version: version,
network: network,
id: fmt.Sprintf("%x", p.ID().Bytes()),
les, les/flowcontrol: improved request serving and flow control (#18230) This change - implements concurrent LES request serving even for a single peer. - replaces the request cost estimation method with a cost table based on benchmarks which gives much more consistent results. Until now the allowed number of light peers was just a guess which probably contributed a lot to the fluctuating quality of available service. Everything related to request cost is implemented in a single object, the 'cost tracker'. It uses a fixed cost table with a global 'correction factor'. Benchmark code is included and can be run at any time to adapt costs to low-level implementation changes. - reimplements flowcontrol.ClientManager in a cleaner and more efficient way, with added capabilities: There is now control over bandwidth, which allows using the flow control parameters for client prioritization. Target utilization over 100 percent is now supported to model concurrent request processing. Total serving bandwidth is reduced during block processing to prevent database contention. - implements an RPC API for the LES servers allowing server operators to assign priority bandwidth to certain clients and change prioritized status even while the client is connected. The new API is meant for cases where server operators charge for LES using an off-protocol mechanism. - adds a unit test for the new client manager. - adds an end-to-end test using the network simulator that tests bandwidth control functions through the new API.
2019-02-26 11:32:48 +00:00
isTrusted: isTrusted,
errCh: make(chan error, 1),
les, les/flowcontrol: improved request serving and flow control (#18230) This change - implements concurrent LES request serving even for a single peer. - replaces the request cost estimation method with a cost table based on benchmarks which gives much more consistent results. Until now the allowed number of light peers was just a guess which probably contributed a lot to the fluctuating quality of available service. Everything related to request cost is implemented in a single object, the 'cost tracker'. It uses a fixed cost table with a global 'correction factor'. Benchmark code is included and can be run at any time to adapt costs to low-level implementation changes. - reimplements flowcontrol.ClientManager in a cleaner and more efficient way, with added capabilities: There is now control over bandwidth, which allows using the flow control parameters for client prioritization. Target utilization over 100 percent is now supported to model concurrent request processing. Total serving bandwidth is reduced during block processing to prevent database contention. - implements an RPC API for the LES servers allowing server operators to assign priority bandwidth to certain clients and change prioritized status even while the client is connected. The new API is meant for cases where server operators charge for LES using an off-protocol mechanism. - adds a unit test for the new client manager. - adds an end-to-end test using the network simulator that tests bandwidth control functions through the new API.
2019-02-26 11:32:48 +00:00
}
}
// rejectUpdate returns true if a parameter update has to be rejected because
// the size and/or rate of updates exceed the capacity limitation
func (p *peer) rejectUpdate(size uint64) bool {
now := mclock.Now()
if p.updateCounter == 0 {
p.updateTime = now
} else {
dt := now - p.updateTime
r := uint64(dt / mclock.AbsTime(allowedUpdateRate))
if p.updateCounter > r {
p.updateCounter -= r
p.updateTime += mclock.AbsTime(allowedUpdateRate * time.Duration(r))
} else {
p.updateCounter = 0
p.updateTime = now
}
2016-10-14 03:51:29 +00:00
}
les, les/flowcontrol: improved request serving and flow control (#18230) This change - implements concurrent LES request serving even for a single peer. - replaces the request cost estimation method with a cost table based on benchmarks which gives much more consistent results. Until now the allowed number of light peers was just a guess which probably contributed a lot to the fluctuating quality of available service. Everything related to request cost is implemented in a single object, the 'cost tracker'. It uses a fixed cost table with a global 'correction factor'. Benchmark code is included and can be run at any time to adapt costs to low-level implementation changes. - reimplements flowcontrol.ClientManager in a cleaner and more efficient way, with added capabilities: There is now control over bandwidth, which allows using the flow control parameters for client prioritization. Target utilization over 100 percent is now supported to model concurrent request processing. Total serving bandwidth is reduced during block processing to prevent database contention. - implements an RPC API for the LES servers allowing server operators to assign priority bandwidth to certain clients and change prioritized status even while the client is connected. The new API is meant for cases where server operators charge for LES using an off-protocol mechanism. - adds a unit test for the new client manager. - adds an end-to-end test using the network simulator that tests bandwidth control functions through the new API.
2019-02-26 11:32:48 +00:00
p.updateCounter += size
return p.updateCounter > allowedUpdateBytes
2016-10-14 03:51:29 +00:00
}
func (p *peer) canQueue() bool {
return p.sendQueue.canQueue()
}
func (p *peer) queueSend(f func()) {
p.sendQueue.queue(f)
}
2016-10-14 03:51:29 +00:00
// Info gathers and returns a collection of metadata known about a peer.
func (p *peer) Info() *eth.PeerInfo {
return &eth.PeerInfo{
Version: p.version,
Difficulty: p.Td(),
Head: fmt.Sprintf("%x", p.Head()),
}
}
// Head retrieves a copy of the current head (most recent) hash of the peer.
func (p *peer) Head() (hash common.Hash) {
p.lock.RLock()
defer p.lock.RUnlock()
copy(hash[:], p.headInfo.Hash[:])
return hash
}
func (p *peer) HeadAndTd() (hash common.Hash, td *big.Int) {
p.lock.RLock()
defer p.lock.RUnlock()
copy(hash[:], p.headInfo.Hash[:])
return hash, p.headInfo.Td
}
func (p *peer) headBlockInfo() blockInfo {
p.lock.RLock()
defer p.lock.RUnlock()
return blockInfo{Hash: p.headInfo.Hash, Number: p.headInfo.Number, Td: p.headInfo.Td}
}
// Td retrieves the current total difficulty of a peer.
func (p *peer) Td() *big.Int {
p.lock.RLock()
defer p.lock.RUnlock()
return new(big.Int).Set(p.headInfo.Td)
}
// waitBefore implements distPeer interface
func (p *peer) waitBefore(maxCost uint64) (time.Duration, float64) {
return p.fcServer.CanSend(maxCost)
}
les, les/flowcontrol: improved request serving and flow control (#18230) This change - implements concurrent LES request serving even for a single peer. - replaces the request cost estimation method with a cost table based on benchmarks which gives much more consistent results. Until now the allowed number of light peers was just a guess which probably contributed a lot to the fluctuating quality of available service. Everything related to request cost is implemented in a single object, the 'cost tracker'. It uses a fixed cost table with a global 'correction factor'. Benchmark code is included and can be run at any time to adapt costs to low-level implementation changes. - reimplements flowcontrol.ClientManager in a cleaner and more efficient way, with added capabilities: There is now control over bandwidth, which allows using the flow control parameters for client prioritization. Target utilization over 100 percent is now supported to model concurrent request processing. Total serving bandwidth is reduced during block processing to prevent database contention. - implements an RPC API for the LES servers allowing server operators to assign priority bandwidth to certain clients and change prioritized status even while the client is connected. The new API is meant for cases where server operators charge for LES using an off-protocol mechanism. - adds a unit test for the new client manager. - adds an end-to-end test using the network simulator that tests bandwidth control functions through the new API.
2019-02-26 11:32:48 +00:00
// updateCapacity updates the request serving capacity assigned to a given client
// and also sends an announcement about the updated flow control parameters
func (p *peer) updateCapacity(cap uint64) {
p.responseLock.Lock()
defer p.responseLock.Unlock()
p.fcParams = flowcontrol.ServerParams{MinRecharge: cap, BufLimit: cap * bufLimitRatio}
p.fcClient.UpdateParams(p.fcParams)
var kvList keyValueList
kvList = kvList.add("flowControl/MRR", cap)
kvList = kvList.add("flowControl/BL", cap*bufLimitRatio)
p.queueSend(func() { p.SendAnnounce(announceData{Update: kvList}) })
}
2016-10-14 03:51:29 +00:00
func sendRequest(w p2p.MsgWriter, msgcode, reqID, cost uint64, data interface{}) error {
type req struct {
ReqID uint64
Data interface{}
}
return p2p.Send(w, msgcode, req{reqID, data})
}
les, les/flowcontrol: improved request serving and flow control (#18230) This change - implements concurrent LES request serving even for a single peer. - replaces the request cost estimation method with a cost table based on benchmarks which gives much more consistent results. Until now the allowed number of light peers was just a guess which probably contributed a lot to the fluctuating quality of available service. Everything related to request cost is implemented in a single object, the 'cost tracker'. It uses a fixed cost table with a global 'correction factor'. Benchmark code is included and can be run at any time to adapt costs to low-level implementation changes. - reimplements flowcontrol.ClientManager in a cleaner and more efficient way, with added capabilities: There is now control over bandwidth, which allows using the flow control parameters for client prioritization. Target utilization over 100 percent is now supported to model concurrent request processing. Total serving bandwidth is reduced during block processing to prevent database contention. - implements an RPC API for the LES servers allowing server operators to assign priority bandwidth to certain clients and change prioritized status even while the client is connected. The new API is meant for cases where server operators charge for LES using an off-protocol mechanism. - adds a unit test for the new client manager. - adds an end-to-end test using the network simulator that tests bandwidth control functions through the new API.
2019-02-26 11:32:48 +00:00
// reply struct represents a reply with the actual data already RLP encoded and
// only the bv (buffer value) missing. This allows the serving mechanism to
// calculate the bv value which depends on the data size before sending the reply.
type reply struct {
w p2p.MsgWriter
msgcode, reqID uint64
data rlp.RawValue
}
// send sends the reply with the calculated buffer value
func (r *reply) send(bv uint64) error {
2016-10-14 03:51:29 +00:00
type resp struct {
ReqID, BV uint64
les, les/flowcontrol: improved request serving and flow control (#18230) This change - implements concurrent LES request serving even for a single peer. - replaces the request cost estimation method with a cost table based on benchmarks which gives much more consistent results. Until now the allowed number of light peers was just a guess which probably contributed a lot to the fluctuating quality of available service. Everything related to request cost is implemented in a single object, the 'cost tracker'. It uses a fixed cost table with a global 'correction factor'. Benchmark code is included and can be run at any time to adapt costs to low-level implementation changes. - reimplements flowcontrol.ClientManager in a cleaner and more efficient way, with added capabilities: There is now control over bandwidth, which allows using the flow control parameters for client prioritization. Target utilization over 100 percent is now supported to model concurrent request processing. Total serving bandwidth is reduced during block processing to prevent database contention. - implements an RPC API for the LES servers allowing server operators to assign priority bandwidth to certain clients and change prioritized status even while the client is connected. The new API is meant for cases where server operators charge for LES using an off-protocol mechanism. - adds a unit test for the new client manager. - adds an end-to-end test using the network simulator that tests bandwidth control functions through the new API.
2019-02-26 11:32:48 +00:00
Data rlp.RawValue
2016-10-14 03:51:29 +00:00
}
les, les/flowcontrol: improved request serving and flow control (#18230) This change - implements concurrent LES request serving even for a single peer. - replaces the request cost estimation method with a cost table based on benchmarks which gives much more consistent results. Until now the allowed number of light peers was just a guess which probably contributed a lot to the fluctuating quality of available service. Everything related to request cost is implemented in a single object, the 'cost tracker'. It uses a fixed cost table with a global 'correction factor'. Benchmark code is included and can be run at any time to adapt costs to low-level implementation changes. - reimplements flowcontrol.ClientManager in a cleaner and more efficient way, with added capabilities: There is now control over bandwidth, which allows using the flow control parameters for client prioritization. Target utilization over 100 percent is now supported to model concurrent request processing. Total serving bandwidth is reduced during block processing to prevent database contention. - implements an RPC API for the LES servers allowing server operators to assign priority bandwidth to certain clients and change prioritized status even while the client is connected. The new API is meant for cases where server operators charge for LES using an off-protocol mechanism. - adds a unit test for the new client manager. - adds an end-to-end test using the network simulator that tests bandwidth control functions through the new API.
2019-02-26 11:32:48 +00:00
return p2p.Send(r.w, r.msgcode, resp{r.reqID, bv, r.data})
}
// size returns the RLP encoded size of the message data
func (r *reply) size() uint32 {
return uint32(len(r.data))
2016-10-14 03:51:29 +00:00
}
func (p *peer) GetRequestCost(msgcode uint64, amount int) uint64 {
p.lock.RLock()
defer p.lock.RUnlock()
2016-10-14 03:51:29 +00:00
cost := p.fcCosts[msgcode].baseCost + p.fcCosts[msgcode].reqCost*uint64(amount)
les, les/flowcontrol: improved request serving and flow control (#18230) This change - implements concurrent LES request serving even for a single peer. - replaces the request cost estimation method with a cost table based on benchmarks which gives much more consistent results. Until now the allowed number of light peers was just a guess which probably contributed a lot to the fluctuating quality of available service. Everything related to request cost is implemented in a single object, the 'cost tracker'. It uses a fixed cost table with a global 'correction factor'. Benchmark code is included and can be run at any time to adapt costs to low-level implementation changes. - reimplements flowcontrol.ClientManager in a cleaner and more efficient way, with added capabilities: There is now control over bandwidth, which allows using the flow control parameters for client prioritization. Target utilization over 100 percent is now supported to model concurrent request processing. Total serving bandwidth is reduced during block processing to prevent database contention. - implements an RPC API for the LES servers allowing server operators to assign priority bandwidth to certain clients and change prioritized status even while the client is connected. The new API is meant for cases where server operators charge for LES using an off-protocol mechanism. - adds a unit test for the new client manager. - adds an end-to-end test using the network simulator that tests bandwidth control functions through the new API.
2019-02-26 11:32:48 +00:00
if cost > p.fcParams.BufLimit {
cost = p.fcParams.BufLimit
}
return cost
}
func (p *peer) GetTxRelayCost(amount, size int) uint64 {
p.lock.RLock()
defer p.lock.RUnlock()
var msgcode uint64
switch p.version {
case lpv1:
msgcode = SendTxMsg
case lpv2:
msgcode = SendTxV2Msg
default:
panic(nil)
}
cost := p.fcCosts[msgcode].baseCost + p.fcCosts[msgcode].reqCost*uint64(amount)
sizeCost := p.fcCosts[msgcode].baseCost + p.fcCosts[msgcode].reqCost*uint64(size)/txSizeCostLimit
if sizeCost > cost {
cost = sizeCost
}
if cost > p.fcParams.BufLimit {
cost = p.fcParams.BufLimit
2016-10-14 03:51:29 +00:00
}
return cost
}
// HasBlock checks if the peer has a given block
func (p *peer) HasBlock(hash common.Hash, number uint64, hasState bool) bool {
p.lock.RLock()
hasBlock := p.hasBlock
p.lock.RUnlock()
return hasBlock != nil && hasBlock(hash, number, hasState)
}
2016-10-14 03:51:29 +00:00
// SendAnnounce announces the availability of a number of blocks through
// a hash notification.
func (p *peer) SendAnnounce(request announceData) error {
return p2p.Send(p.rw, AnnounceMsg, request)
}
les, les/flowcontrol: improved request serving and flow control (#18230) This change - implements concurrent LES request serving even for a single peer. - replaces the request cost estimation method with a cost table based on benchmarks which gives much more consistent results. Until now the allowed number of light peers was just a guess which probably contributed a lot to the fluctuating quality of available service. Everything related to request cost is implemented in a single object, the 'cost tracker'. It uses a fixed cost table with a global 'correction factor'. Benchmark code is included and can be run at any time to adapt costs to low-level implementation changes. - reimplements flowcontrol.ClientManager in a cleaner and more efficient way, with added capabilities: There is now control over bandwidth, which allows using the flow control parameters for client prioritization. Target utilization over 100 percent is now supported to model concurrent request processing. Total serving bandwidth is reduced during block processing to prevent database contention. - implements an RPC API for the LES servers allowing server operators to assign priority bandwidth to certain clients and change prioritized status even while the client is connected. The new API is meant for cases where server operators charge for LES using an off-protocol mechanism. - adds a unit test for the new client manager. - adds an end-to-end test using the network simulator that tests bandwidth control functions through the new API.
2019-02-26 11:32:48 +00:00
// ReplyBlockHeaders creates a reply with a batch of block headers
func (p *peer) ReplyBlockHeaders(reqID uint64, headers []*types.Header) *reply {
data, _ := rlp.EncodeToBytes(headers)
return &reply{p.rw, BlockHeadersMsg, reqID, data}
2016-10-14 03:51:29 +00:00
}
les, les/flowcontrol: improved request serving and flow control (#18230) This change - implements concurrent LES request serving even for a single peer. - replaces the request cost estimation method with a cost table based on benchmarks which gives much more consistent results. Until now the allowed number of light peers was just a guess which probably contributed a lot to the fluctuating quality of available service. Everything related to request cost is implemented in a single object, the 'cost tracker'. It uses a fixed cost table with a global 'correction factor'. Benchmark code is included and can be run at any time to adapt costs to low-level implementation changes. - reimplements flowcontrol.ClientManager in a cleaner and more efficient way, with added capabilities: There is now control over bandwidth, which allows using the flow control parameters for client prioritization. Target utilization over 100 percent is now supported to model concurrent request processing. Total serving bandwidth is reduced during block processing to prevent database contention. - implements an RPC API for the LES servers allowing server operators to assign priority bandwidth to certain clients and change prioritized status even while the client is connected. The new API is meant for cases where server operators charge for LES using an off-protocol mechanism. - adds a unit test for the new client manager. - adds an end-to-end test using the network simulator that tests bandwidth control functions through the new API.
2019-02-26 11:32:48 +00:00
// ReplyBlockBodiesRLP creates a reply with a batch of block contents from
2016-10-14 03:51:29 +00:00
// an already RLP encoded format.
les, les/flowcontrol: improved request serving and flow control (#18230) This change - implements concurrent LES request serving even for a single peer. - replaces the request cost estimation method with a cost table based on benchmarks which gives much more consistent results. Until now the allowed number of light peers was just a guess which probably contributed a lot to the fluctuating quality of available service. Everything related to request cost is implemented in a single object, the 'cost tracker'. It uses a fixed cost table with a global 'correction factor'. Benchmark code is included and can be run at any time to adapt costs to low-level implementation changes. - reimplements flowcontrol.ClientManager in a cleaner and more efficient way, with added capabilities: There is now control over bandwidth, which allows using the flow control parameters for client prioritization. Target utilization over 100 percent is now supported to model concurrent request processing. Total serving bandwidth is reduced during block processing to prevent database contention. - implements an RPC API for the LES servers allowing server operators to assign priority bandwidth to certain clients and change prioritized status even while the client is connected. The new API is meant for cases where server operators charge for LES using an off-protocol mechanism. - adds a unit test for the new client manager. - adds an end-to-end test using the network simulator that tests bandwidth control functions through the new API.
2019-02-26 11:32:48 +00:00
func (p *peer) ReplyBlockBodiesRLP(reqID uint64, bodies []rlp.RawValue) *reply {
data, _ := rlp.EncodeToBytes(bodies)
return &reply{p.rw, BlockBodiesMsg, reqID, data}
2016-10-14 03:51:29 +00:00
}
les, les/flowcontrol: improved request serving and flow control (#18230) This change - implements concurrent LES request serving even for a single peer. - replaces the request cost estimation method with a cost table based on benchmarks which gives much more consistent results. Until now the allowed number of light peers was just a guess which probably contributed a lot to the fluctuating quality of available service. Everything related to request cost is implemented in a single object, the 'cost tracker'. It uses a fixed cost table with a global 'correction factor'. Benchmark code is included and can be run at any time to adapt costs to low-level implementation changes. - reimplements flowcontrol.ClientManager in a cleaner and more efficient way, with added capabilities: There is now control over bandwidth, which allows using the flow control parameters for client prioritization. Target utilization over 100 percent is now supported to model concurrent request processing. Total serving bandwidth is reduced during block processing to prevent database contention. - implements an RPC API for the LES servers allowing server operators to assign priority bandwidth to certain clients and change prioritized status even while the client is connected. The new API is meant for cases where server operators charge for LES using an off-protocol mechanism. - adds a unit test for the new client manager. - adds an end-to-end test using the network simulator that tests bandwidth control functions through the new API.
2019-02-26 11:32:48 +00:00
// ReplyCode creates a reply with a batch of arbitrary internal data, corresponding to the
2016-10-14 03:51:29 +00:00
// hashes requested.
les, les/flowcontrol: improved request serving and flow control (#18230) This change - implements concurrent LES request serving even for a single peer. - replaces the request cost estimation method with a cost table based on benchmarks which gives much more consistent results. Until now the allowed number of light peers was just a guess which probably contributed a lot to the fluctuating quality of available service. Everything related to request cost is implemented in a single object, the 'cost tracker'. It uses a fixed cost table with a global 'correction factor'. Benchmark code is included and can be run at any time to adapt costs to low-level implementation changes. - reimplements flowcontrol.ClientManager in a cleaner and more efficient way, with added capabilities: There is now control over bandwidth, which allows using the flow control parameters for client prioritization. Target utilization over 100 percent is now supported to model concurrent request processing. Total serving bandwidth is reduced during block processing to prevent database contention. - implements an RPC API for the LES servers allowing server operators to assign priority bandwidth to certain clients and change prioritized status even while the client is connected. The new API is meant for cases where server operators charge for LES using an off-protocol mechanism. - adds a unit test for the new client manager. - adds an end-to-end test using the network simulator that tests bandwidth control functions through the new API.
2019-02-26 11:32:48 +00:00
func (p *peer) ReplyCode(reqID uint64, codes [][]byte) *reply {
data, _ := rlp.EncodeToBytes(codes)
return &reply{p.rw, CodeMsg, reqID, data}
2016-10-14 03:51:29 +00:00
}
les, les/flowcontrol: improved request serving and flow control (#18230) This change - implements concurrent LES request serving even for a single peer. - replaces the request cost estimation method with a cost table based on benchmarks which gives much more consistent results. Until now the allowed number of light peers was just a guess which probably contributed a lot to the fluctuating quality of available service. Everything related to request cost is implemented in a single object, the 'cost tracker'. It uses a fixed cost table with a global 'correction factor'. Benchmark code is included and can be run at any time to adapt costs to low-level implementation changes. - reimplements flowcontrol.ClientManager in a cleaner and more efficient way, with added capabilities: There is now control over bandwidth, which allows using the flow control parameters for client prioritization. Target utilization over 100 percent is now supported to model concurrent request processing. Total serving bandwidth is reduced during block processing to prevent database contention. - implements an RPC API for the LES servers allowing server operators to assign priority bandwidth to certain clients and change prioritized status even while the client is connected. The new API is meant for cases where server operators charge for LES using an off-protocol mechanism. - adds a unit test for the new client manager. - adds an end-to-end test using the network simulator that tests bandwidth control functions through the new API.
2019-02-26 11:32:48 +00:00
// ReplyReceiptsRLP creates a reply with a batch of transaction receipts, corresponding to the
2016-10-14 03:51:29 +00:00
// ones requested from an already RLP encoded format.
les, les/flowcontrol: improved request serving and flow control (#18230) This change - implements concurrent LES request serving even for a single peer. - replaces the request cost estimation method with a cost table based on benchmarks which gives much more consistent results. Until now the allowed number of light peers was just a guess which probably contributed a lot to the fluctuating quality of available service. Everything related to request cost is implemented in a single object, the 'cost tracker'. It uses a fixed cost table with a global 'correction factor'. Benchmark code is included and can be run at any time to adapt costs to low-level implementation changes. - reimplements flowcontrol.ClientManager in a cleaner and more efficient way, with added capabilities: There is now control over bandwidth, which allows using the flow control parameters for client prioritization. Target utilization over 100 percent is now supported to model concurrent request processing. Total serving bandwidth is reduced during block processing to prevent database contention. - implements an RPC API for the LES servers allowing server operators to assign priority bandwidth to certain clients and change prioritized status even while the client is connected. The new API is meant for cases where server operators charge for LES using an off-protocol mechanism. - adds a unit test for the new client manager. - adds an end-to-end test using the network simulator that tests bandwidth control functions through the new API.
2019-02-26 11:32:48 +00:00
func (p *peer) ReplyReceiptsRLP(reqID uint64, receipts []rlp.RawValue) *reply {
data, _ := rlp.EncodeToBytes(receipts)
return &reply{p.rw, ReceiptsMsg, reqID, data}
2016-10-14 03:51:29 +00:00
}
les, les/flowcontrol: improved request serving and flow control (#18230) This change - implements concurrent LES request serving even for a single peer. - replaces the request cost estimation method with a cost table based on benchmarks which gives much more consistent results. Until now the allowed number of light peers was just a guess which probably contributed a lot to the fluctuating quality of available service. Everything related to request cost is implemented in a single object, the 'cost tracker'. It uses a fixed cost table with a global 'correction factor'. Benchmark code is included and can be run at any time to adapt costs to low-level implementation changes. - reimplements flowcontrol.ClientManager in a cleaner and more efficient way, with added capabilities: There is now control over bandwidth, which allows using the flow control parameters for client prioritization. Target utilization over 100 percent is now supported to model concurrent request processing. Total serving bandwidth is reduced during block processing to prevent database contention. - implements an RPC API for the LES servers allowing server operators to assign priority bandwidth to certain clients and change prioritized status even while the client is connected. The new API is meant for cases where server operators charge for LES using an off-protocol mechanism. - adds a unit test for the new client manager. - adds an end-to-end test using the network simulator that tests bandwidth control functions through the new API.
2019-02-26 11:32:48 +00:00
// ReplyProofs creates a reply with a batch of legacy LES/1 merkle proofs, corresponding to the ones requested.
func (p *peer) ReplyProofs(reqID uint64, proofs proofsData) *reply {
data, _ := rlp.EncodeToBytes(proofs)
return &reply{p.rw, ProofsV1Msg, reqID, data}
2016-10-14 03:51:29 +00:00
}
les, les/flowcontrol: improved request serving and flow control (#18230) This change - implements concurrent LES request serving even for a single peer. - replaces the request cost estimation method with a cost table based on benchmarks which gives much more consistent results. Until now the allowed number of light peers was just a guess which probably contributed a lot to the fluctuating quality of available service. Everything related to request cost is implemented in a single object, the 'cost tracker'. It uses a fixed cost table with a global 'correction factor'. Benchmark code is included and can be run at any time to adapt costs to low-level implementation changes. - reimplements flowcontrol.ClientManager in a cleaner and more efficient way, with added capabilities: There is now control over bandwidth, which allows using the flow control parameters for client prioritization. Target utilization over 100 percent is now supported to model concurrent request processing. Total serving bandwidth is reduced during block processing to prevent database contention. - implements an RPC API for the LES servers allowing server operators to assign priority bandwidth to certain clients and change prioritized status even while the client is connected. The new API is meant for cases where server operators charge for LES using an off-protocol mechanism. - adds a unit test for the new client manager. - adds an end-to-end test using the network simulator that tests bandwidth control functions through the new API.
2019-02-26 11:32:48 +00:00
// ReplyProofsV2 creates a reply with a batch of merkle proofs, corresponding to the ones requested.
func (p *peer) ReplyProofsV2(reqID uint64, proofs light.NodeList) *reply {
data, _ := rlp.EncodeToBytes(proofs)
return &reply{p.rw, ProofsV2Msg, reqID, data}
}
les, les/flowcontrol: improved request serving and flow control (#18230) This change - implements concurrent LES request serving even for a single peer. - replaces the request cost estimation method with a cost table based on benchmarks which gives much more consistent results. Until now the allowed number of light peers was just a guess which probably contributed a lot to the fluctuating quality of available service. Everything related to request cost is implemented in a single object, the 'cost tracker'. It uses a fixed cost table with a global 'correction factor'. Benchmark code is included and can be run at any time to adapt costs to low-level implementation changes. - reimplements flowcontrol.ClientManager in a cleaner and more efficient way, with added capabilities: There is now control over bandwidth, which allows using the flow control parameters for client prioritization. Target utilization over 100 percent is now supported to model concurrent request processing. Total serving bandwidth is reduced during block processing to prevent database contention. - implements an RPC API for the LES servers allowing server operators to assign priority bandwidth to certain clients and change prioritized status even while the client is connected. The new API is meant for cases where server operators charge for LES using an off-protocol mechanism. - adds a unit test for the new client manager. - adds an end-to-end test using the network simulator that tests bandwidth control functions through the new API.
2019-02-26 11:32:48 +00:00
// ReplyHeaderProofs creates a reply with a batch of legacy LES/1 header proofs, corresponding to the ones requested.
func (p *peer) ReplyHeaderProofs(reqID uint64, proofs []ChtResp) *reply {
data, _ := rlp.EncodeToBytes(proofs)
return &reply{p.rw, HeaderProofsMsg, reqID, data}
2016-10-14 03:51:29 +00:00
}
les, les/flowcontrol: improved request serving and flow control (#18230) This change - implements concurrent LES request serving even for a single peer. - replaces the request cost estimation method with a cost table based on benchmarks which gives much more consistent results. Until now the allowed number of light peers was just a guess which probably contributed a lot to the fluctuating quality of available service. Everything related to request cost is implemented in a single object, the 'cost tracker'. It uses a fixed cost table with a global 'correction factor'. Benchmark code is included and can be run at any time to adapt costs to low-level implementation changes. - reimplements flowcontrol.ClientManager in a cleaner and more efficient way, with added capabilities: There is now control over bandwidth, which allows using the flow control parameters for client prioritization. Target utilization over 100 percent is now supported to model concurrent request processing. Total serving bandwidth is reduced during block processing to prevent database contention. - implements an RPC API for the LES servers allowing server operators to assign priority bandwidth to certain clients and change prioritized status even while the client is connected. The new API is meant for cases where server operators charge for LES using an off-protocol mechanism. - adds a unit test for the new client manager. - adds an end-to-end test using the network simulator that tests bandwidth control functions through the new API.
2019-02-26 11:32:48 +00:00
// ReplyHelperTrieProofs creates a reply with a batch of HelperTrie proofs, corresponding to the ones requested.
func (p *peer) ReplyHelperTrieProofs(reqID uint64, resp HelperTrieResps) *reply {
data, _ := rlp.EncodeToBytes(resp)
return &reply{p.rw, HelperTrieProofsMsg, reqID, data}
}
les, les/flowcontrol: improved request serving and flow control (#18230) This change - implements concurrent LES request serving even for a single peer. - replaces the request cost estimation method with a cost table based on benchmarks which gives much more consistent results. Until now the allowed number of light peers was just a guess which probably contributed a lot to the fluctuating quality of available service. Everything related to request cost is implemented in a single object, the 'cost tracker'. It uses a fixed cost table with a global 'correction factor'. Benchmark code is included and can be run at any time to adapt costs to low-level implementation changes. - reimplements flowcontrol.ClientManager in a cleaner and more efficient way, with added capabilities: There is now control over bandwidth, which allows using the flow control parameters for client prioritization. Target utilization over 100 percent is now supported to model concurrent request processing. Total serving bandwidth is reduced during block processing to prevent database contention. - implements an RPC API for the LES servers allowing server operators to assign priority bandwidth to certain clients and change prioritized status even while the client is connected. The new API is meant for cases where server operators charge for LES using an off-protocol mechanism. - adds a unit test for the new client manager. - adds an end-to-end test using the network simulator that tests bandwidth control functions through the new API.
2019-02-26 11:32:48 +00:00
// ReplyTxStatus creates a reply with a batch of transaction status records, corresponding to the ones requested.
func (p *peer) ReplyTxStatus(reqID uint64, stats []txStatus) *reply {
data, _ := rlp.EncodeToBytes(stats)
return &reply{p.rw, TxStatusMsg, reqID, data}
}
2016-10-14 03:51:29 +00:00
// RequestHeadersByHash fetches a batch of blocks' headers corresponding to the
// specified header query, based on the hash of an origin block.
func (p *peer) RequestHeadersByHash(reqID, cost uint64, origin common.Hash, amount int, skip int, reverse bool) error {
p.Log().Debug("Fetching batch of headers", "count", amount, "fromhash", origin, "skip", skip, "reverse", reverse)
2016-10-14 03:51:29 +00:00
return sendRequest(p.rw, GetBlockHeadersMsg, reqID, cost, &getBlockHeadersData{Origin: hashOrNumber{Hash: origin}, Amount: uint64(amount), Skip: uint64(skip), Reverse: reverse})
}
// RequestHeadersByNumber fetches a batch of blocks' headers corresponding to the
// specified header query, based on the number of an origin block.
func (p *peer) RequestHeadersByNumber(reqID, cost, origin uint64, amount int, skip int, reverse bool) error {
p.Log().Debug("Fetching batch of headers", "count", amount, "fromnum", origin, "skip", skip, "reverse", reverse)
2016-10-14 03:51:29 +00:00
return sendRequest(p.rw, GetBlockHeadersMsg, reqID, cost, &getBlockHeadersData{Origin: hashOrNumber{Number: origin}, Amount: uint64(amount), Skip: uint64(skip), Reverse: reverse})
}
// RequestBodies fetches a batch of blocks' bodies corresponding to the hashes
// specified.
func (p *peer) RequestBodies(reqID, cost uint64, hashes []common.Hash) error {
p.Log().Debug("Fetching batch of block bodies", "count", len(hashes))
2016-10-14 03:51:29 +00:00
return sendRequest(p.rw, GetBlockBodiesMsg, reqID, cost, hashes)
}
// RequestCode fetches a batch of arbitrary data from a node's known state
// data, corresponding to the specified hashes.
func (p *peer) RequestCode(reqID, cost uint64, reqs []CodeReq) error {
p.Log().Debug("Fetching batch of codes", "count", len(reqs))
2016-10-14 03:51:29 +00:00
return sendRequest(p.rw, GetCodeMsg, reqID, cost, reqs)
}
// RequestReceipts fetches a batch of transaction receipts from a remote node.
func (p *peer) RequestReceipts(reqID, cost uint64, hashes []common.Hash) error {
p.Log().Debug("Fetching batch of receipts", "count", len(hashes))
2016-10-14 03:51:29 +00:00
return sendRequest(p.rw, GetReceiptsMsg, reqID, cost, hashes)
}
// RequestProofs fetches a batch of merkle proofs from a remote node.
func (p *peer) RequestProofs(reqID, cost uint64, reqs []ProofReq) error {
p.Log().Debug("Fetching batch of proofs", "count", len(reqs))
switch p.version {
case lpv1:
return sendRequest(p.rw, GetProofsV1Msg, reqID, cost, reqs)
case lpv2:
return sendRequest(p.rw, GetProofsV2Msg, reqID, cost, reqs)
default:
panic(nil)
}
}
// RequestHelperTrieProofs fetches a batch of HelperTrie merkle proofs from a remote node.
func (p *peer) RequestHelperTrieProofs(reqID, cost uint64, data interface{}) error {
switch p.version {
case lpv1:
reqs, ok := data.([]ChtReq)
if !ok {
return errInvalidHelpTrieReq
}
p.Log().Debug("Fetching batch of header proofs", "count", len(reqs))
return sendRequest(p.rw, GetHeaderProofsMsg, reqID, cost, reqs)
case lpv2:
reqs, ok := data.([]HelperTrieReq)
if !ok {
return errInvalidHelpTrieReq
}
p.Log().Debug("Fetching batch of HelperTrie proofs", "count", len(reqs))
return sendRequest(p.rw, GetHelperTrieProofsMsg, reqID, cost, reqs)
default:
panic(nil)
}
2016-10-14 03:51:29 +00:00
}
// RequestTxStatus fetches a batch of transaction status records from a remote node.
func (p *peer) RequestTxStatus(reqID, cost uint64, txHashes []common.Hash) error {
p.Log().Debug("Requesting transaction status", "count", len(txHashes))
return sendRequest(p.rw, GetTxStatusMsg, reqID, cost, txHashes)
2016-10-14 03:51:29 +00:00
}
les, les/flowcontrol: improved request serving and flow control (#18230) This change - implements concurrent LES request serving even for a single peer. - replaces the request cost estimation method with a cost table based on benchmarks which gives much more consistent results. Until now the allowed number of light peers was just a guess which probably contributed a lot to the fluctuating quality of available service. Everything related to request cost is implemented in a single object, the 'cost tracker'. It uses a fixed cost table with a global 'correction factor'. Benchmark code is included and can be run at any time to adapt costs to low-level implementation changes. - reimplements flowcontrol.ClientManager in a cleaner and more efficient way, with added capabilities: There is now control over bandwidth, which allows using the flow control parameters for client prioritization. Target utilization over 100 percent is now supported to model concurrent request processing. Total serving bandwidth is reduced during block processing to prevent database contention. - implements an RPC API for the LES servers allowing server operators to assign priority bandwidth to certain clients and change prioritized status even while the client is connected. The new API is meant for cases where server operators charge for LES using an off-protocol mechanism. - adds a unit test for the new client manager. - adds an end-to-end test using the network simulator that tests bandwidth control functions through the new API.
2019-02-26 11:32:48 +00:00
// SendTxStatus creates a reply with a batch of transactions to be added to the remote transaction pool.
func (p *peer) SendTxs(reqID, cost uint64, txs rlp.RawValue) error {
p.Log().Debug("Sending batch of transactions", "size", len(txs))
switch p.version {
case lpv1:
return p2p.Send(p.rw, SendTxMsg, txs) // old message format does not include reqID
case lpv2:
return sendRequest(p.rw, SendTxV2Msg, reqID, cost, txs)
default:
panic(nil)
}
2016-10-14 03:51:29 +00:00
}
type keyValueEntry struct {
Key string
Value rlp.RawValue
}
type keyValueList []keyValueEntry
type keyValueMap map[string]rlp.RawValue
func (l keyValueList) add(key string, val interface{}) keyValueList {
var entry keyValueEntry
entry.Key = key
if val == nil {
val = uint64(0)
}
enc, err := rlp.EncodeToBytes(val)
if err == nil {
entry.Value = enc
}
return append(l, entry)
}
les, les/flowcontrol: improved request serving and flow control (#18230) This change - implements concurrent LES request serving even for a single peer. - replaces the request cost estimation method with a cost table based on benchmarks which gives much more consistent results. Until now the allowed number of light peers was just a guess which probably contributed a lot to the fluctuating quality of available service. Everything related to request cost is implemented in a single object, the 'cost tracker'. It uses a fixed cost table with a global 'correction factor'. Benchmark code is included and can be run at any time to adapt costs to low-level implementation changes. - reimplements flowcontrol.ClientManager in a cleaner and more efficient way, with added capabilities: There is now control over bandwidth, which allows using the flow control parameters for client prioritization. Target utilization over 100 percent is now supported to model concurrent request processing. Total serving bandwidth is reduced during block processing to prevent database contention. - implements an RPC API for the LES servers allowing server operators to assign priority bandwidth to certain clients and change prioritized status even while the client is connected. The new API is meant for cases where server operators charge for LES using an off-protocol mechanism. - adds a unit test for the new client manager. - adds an end-to-end test using the network simulator that tests bandwidth control functions through the new API.
2019-02-26 11:32:48 +00:00
func (l keyValueList) decode() (keyValueMap, uint64) {
2016-10-14 03:51:29 +00:00
m := make(keyValueMap)
les, les/flowcontrol: improved request serving and flow control (#18230) This change - implements concurrent LES request serving even for a single peer. - replaces the request cost estimation method with a cost table based on benchmarks which gives much more consistent results. Until now the allowed number of light peers was just a guess which probably contributed a lot to the fluctuating quality of available service. Everything related to request cost is implemented in a single object, the 'cost tracker'. It uses a fixed cost table with a global 'correction factor'. Benchmark code is included and can be run at any time to adapt costs to low-level implementation changes. - reimplements flowcontrol.ClientManager in a cleaner and more efficient way, with added capabilities: There is now control over bandwidth, which allows using the flow control parameters for client prioritization. Target utilization over 100 percent is now supported to model concurrent request processing. Total serving bandwidth is reduced during block processing to prevent database contention. - implements an RPC API for the LES servers allowing server operators to assign priority bandwidth to certain clients and change prioritized status even while the client is connected. The new API is meant for cases where server operators charge for LES using an off-protocol mechanism. - adds a unit test for the new client manager. - adds an end-to-end test using the network simulator that tests bandwidth control functions through the new API.
2019-02-26 11:32:48 +00:00
var size uint64
2016-10-14 03:51:29 +00:00
for _, entry := range l {
m[entry.Key] = entry.Value
les, les/flowcontrol: improved request serving and flow control (#18230) This change - implements concurrent LES request serving even for a single peer. - replaces the request cost estimation method with a cost table based on benchmarks which gives much more consistent results. Until now the allowed number of light peers was just a guess which probably contributed a lot to the fluctuating quality of available service. Everything related to request cost is implemented in a single object, the 'cost tracker'. It uses a fixed cost table with a global 'correction factor'. Benchmark code is included and can be run at any time to adapt costs to low-level implementation changes. - reimplements flowcontrol.ClientManager in a cleaner and more efficient way, with added capabilities: There is now control over bandwidth, which allows using the flow control parameters for client prioritization. Target utilization over 100 percent is now supported to model concurrent request processing. Total serving bandwidth is reduced during block processing to prevent database contention. - implements an RPC API for the LES servers allowing server operators to assign priority bandwidth to certain clients and change prioritized status even while the client is connected. The new API is meant for cases where server operators charge for LES using an off-protocol mechanism. - adds a unit test for the new client manager. - adds an end-to-end test using the network simulator that tests bandwidth control functions through the new API.
2019-02-26 11:32:48 +00:00
size += uint64(len(entry.Key)) + uint64(len(entry.Value)) + 8
2016-10-14 03:51:29 +00:00
}
les, les/flowcontrol: improved request serving and flow control (#18230) This change - implements concurrent LES request serving even for a single peer. - replaces the request cost estimation method with a cost table based on benchmarks which gives much more consistent results. Until now the allowed number of light peers was just a guess which probably contributed a lot to the fluctuating quality of available service. Everything related to request cost is implemented in a single object, the 'cost tracker'. It uses a fixed cost table with a global 'correction factor'. Benchmark code is included and can be run at any time to adapt costs to low-level implementation changes. - reimplements flowcontrol.ClientManager in a cleaner and more efficient way, with added capabilities: There is now control over bandwidth, which allows using the flow control parameters for client prioritization. Target utilization over 100 percent is now supported to model concurrent request processing. Total serving bandwidth is reduced during block processing to prevent database contention. - implements an RPC API for the LES servers allowing server operators to assign priority bandwidth to certain clients and change prioritized status even while the client is connected. The new API is meant for cases where server operators charge for LES using an off-protocol mechanism. - adds a unit test for the new client manager. - adds an end-to-end test using the network simulator that tests bandwidth control functions through the new API.
2019-02-26 11:32:48 +00:00
return m, size
2016-10-14 03:51:29 +00:00
}
func (m keyValueMap) get(key string, val interface{}) error {
enc, ok := m[key]
if !ok {
return errResp(ErrMissingKey, "%s", key)
2016-10-14 03:51:29 +00:00
}
if val == nil {
return nil
}
return rlp.DecodeBytes(enc, val)
}
func (p *peer) sendReceiveHandshake(sendList keyValueList) (keyValueList, error) {
// Send out own handshake in a new thread
errc := make(chan error, 1)
go func() {
errc <- p2p.Send(p.rw, StatusMsg, sendList)
}()
// In the mean time retrieve the remote status message
msg, err := p.rw.ReadMsg()
if err != nil {
return nil, err
}
if msg.Code != StatusMsg {
return nil, errResp(ErrNoStatusMsg, "first msg has code %x (!= %x)", msg.Code, StatusMsg)
}
if msg.Size > ProtocolMaxMsgSize {
return nil, errResp(ErrMsgTooLarge, "%v > %v", msg.Size, ProtocolMaxMsgSize)
}
// Decode the handshake
var recvList keyValueList
if err := msg.Decode(&recvList); err != nil {
return nil, errResp(ErrDecode, "msg %v: %v", msg, err)
}
if err := <-errc; err != nil {
return nil, err
}
return recvList, nil
}
// Handshake executes the les protocol handshake, negotiating version number,
// network IDs, difficulties, head and genesis blocks.
func (p *peer) Handshake(td *big.Int, head common.Hash, headNum uint64, genesis common.Hash, server *LesServer) error {
p.lock.Lock()
defer p.lock.Unlock()
var send keyValueList
send = send.add("protocolVersion", uint64(p.version))
send = send.add("networkId", p.network)
2016-10-14 03:51:29 +00:00
send = send.add("headTd", td)
send = send.add("headHash", head)
send = send.add("headNum", headNum)
send = send.add("genesisHash", genesis)
if server != nil {
if !server.onlyAnnounce {
//only announce server. It sends only announse requests
send = send.add("serveHeaders", nil)
send = send.add("serveChainSince", uint64(0))
send = send.add("serveStateSince", uint64(0))
send = send.add("txRelay", nil)
}
2016-10-14 03:51:29 +00:00
send = send.add("flowControl/BL", server.defParams.BufLimit)
send = send.add("flowControl/MRR", server.defParams.MinRecharge)
les, les/flowcontrol: improved request serving and flow control (#18230) This change - implements concurrent LES request serving even for a single peer. - replaces the request cost estimation method with a cost table based on benchmarks which gives much more consistent results. Until now the allowed number of light peers was just a guess which probably contributed a lot to the fluctuating quality of available service. Everything related to request cost is implemented in a single object, the 'cost tracker'. It uses a fixed cost table with a global 'correction factor'. Benchmark code is included and can be run at any time to adapt costs to low-level implementation changes. - reimplements flowcontrol.ClientManager in a cleaner and more efficient way, with added capabilities: There is now control over bandwidth, which allows using the flow control parameters for client prioritization. Target utilization over 100 percent is now supported to model concurrent request processing. Total serving bandwidth is reduced during block processing to prevent database contention. - implements an RPC API for the LES servers allowing server operators to assign priority bandwidth to certain clients and change prioritized status even while the client is connected. The new API is meant for cases where server operators charge for LES using an off-protocol mechanism. - adds a unit test for the new client manager. - adds an end-to-end test using the network simulator that tests bandwidth control functions through the new API.
2019-02-26 11:32:48 +00:00
var costList RequestCostList
if server.costTracker != nil {
costList = server.costTracker.makeCostList()
} else {
costList = testCostList()
}
send = send.add("flowControl/MRC", costList)
p.fcCosts = costList.decode()
p.fcParams = server.defParams
} else {
//on client node
p.announceType = announceTypeSimple
if p.isTrusted {
p.announceType = announceTypeSigned
}
send = send.add("announceType", p.announceType)
2016-10-14 03:51:29 +00:00
}
2016-10-14 03:51:29 +00:00
recvList, err := p.sendReceiveHandshake(send)
if err != nil {
return err
}
les, les/flowcontrol: improved request serving and flow control (#18230) This change - implements concurrent LES request serving even for a single peer. - replaces the request cost estimation method with a cost table based on benchmarks which gives much more consistent results. Until now the allowed number of light peers was just a guess which probably contributed a lot to the fluctuating quality of available service. Everything related to request cost is implemented in a single object, the 'cost tracker'. It uses a fixed cost table with a global 'correction factor'. Benchmark code is included and can be run at any time to adapt costs to low-level implementation changes. - reimplements flowcontrol.ClientManager in a cleaner and more efficient way, with added capabilities: There is now control over bandwidth, which allows using the flow control parameters for client prioritization. Target utilization over 100 percent is now supported to model concurrent request processing. Total serving bandwidth is reduced during block processing to prevent database contention. - implements an RPC API for the LES servers allowing server operators to assign priority bandwidth to certain clients and change prioritized status even while the client is connected. The new API is meant for cases where server operators charge for LES using an off-protocol mechanism. - adds a unit test for the new client manager. - adds an end-to-end test using the network simulator that tests bandwidth control functions through the new API.
2019-02-26 11:32:48 +00:00
recv, size := recvList.decode()
if p.rejectUpdate(size) {
return errResp(ErrRequestRejected, "")
}
2016-10-14 03:51:29 +00:00
var rGenesis, rHash common.Hash
var rVersion, rNetwork, rNum uint64
var rTd *big.Int
if err := recv.get("protocolVersion", &rVersion); err != nil {
return err
}
if err := recv.get("networkId", &rNetwork); err != nil {
return err
}
if err := recv.get("headTd", &rTd); err != nil {
return err
}
if err := recv.get("headHash", &rHash); err != nil {
return err
}
if err := recv.get("headNum", &rNum); err != nil {
return err
}
if err := recv.get("genesisHash", &rGenesis); err != nil {
return err
}
if rGenesis != genesis {
return errResp(ErrGenesisBlockMismatch, "%x (!= %x)", rGenesis[:8], genesis[:8])
2016-10-14 03:51:29 +00:00
}
if rNetwork != p.network {
2016-10-14 03:51:29 +00:00
return errResp(ErrNetworkIdMismatch, "%d (!= %d)", rNetwork, p.network)
}
if int(rVersion) != p.version {
return errResp(ErrProtocolVersionMismatch, "%d (!= %d)", rVersion, p.version)
}
2016-10-14 03:51:29 +00:00
if server != nil {
// until we have a proper peer connectivity API, allow LES connection to other servers
/*if recv.get("serveStateSince", nil) == nil {
2016-10-14 03:51:29 +00:00
return errResp(ErrUselessPeer, "wanted client, got server")
}*/
if recv.get("announceType", &p.announceType) != nil {
//set default announceType on server side
p.announceType = announceTypeSimple
}
2016-10-14 03:51:29 +00:00
p.fcClient = flowcontrol.NewClientNode(server.fcManager, server.defParams)
} else {
//mark OnlyAnnounce server if "serveHeaders", "serveChainSince", "serveStateSince" or "txRelay" fields don't exist
2016-10-14 03:51:29 +00:00
if recv.get("serveChainSince", nil) != nil {
p.isOnlyAnnounce = true
2016-10-14 03:51:29 +00:00
}
if recv.get("serveStateSince", nil) != nil {
p.isOnlyAnnounce = true
2016-10-14 03:51:29 +00:00
}
if recv.get("txRelay", nil) != nil {
p.isOnlyAnnounce = true
2016-10-14 03:51:29 +00:00
}
if p.isOnlyAnnounce && !p.isTrusted {
return errResp(ErrUselessPeer, "peer cannot serve requests")
}
les, les/flowcontrol: improved request serving and flow control (#18230) This change - implements concurrent LES request serving even for a single peer. - replaces the request cost estimation method with a cost table based on benchmarks which gives much more consistent results. Until now the allowed number of light peers was just a guess which probably contributed a lot to the fluctuating quality of available service. Everything related to request cost is implemented in a single object, the 'cost tracker'. It uses a fixed cost table with a global 'correction factor'. Benchmark code is included and can be run at any time to adapt costs to low-level implementation changes. - reimplements flowcontrol.ClientManager in a cleaner and more efficient way, with added capabilities: There is now control over bandwidth, which allows using the flow control parameters for client prioritization. Target utilization over 100 percent is now supported to model concurrent request processing. Total serving bandwidth is reduced during block processing to prevent database contention. - implements an RPC API for the LES servers allowing server operators to assign priority bandwidth to certain clients and change prioritized status even while the client is connected. The new API is meant for cases where server operators charge for LES using an off-protocol mechanism. - adds a unit test for the new client manager. - adds an end-to-end test using the network simulator that tests bandwidth control functions through the new API.
2019-02-26 11:32:48 +00:00
var params flowcontrol.ServerParams
2016-10-14 03:51:29 +00:00
if err := recv.get("flowControl/BL", &params.BufLimit); err != nil {
return err
}
if err := recv.get("flowControl/MRR", &params.MinRecharge); err != nil {
return err
}
var MRC RequestCostList
if err := recv.get("flowControl/MRC", &MRC); err != nil {
return err
}
les, les/flowcontrol: improved request serving and flow control (#18230) This change - implements concurrent LES request serving even for a single peer. - replaces the request cost estimation method with a cost table based on benchmarks which gives much more consistent results. Until now the allowed number of light peers was just a guess which probably contributed a lot to the fluctuating quality of available service. Everything related to request cost is implemented in a single object, the 'cost tracker'. It uses a fixed cost table with a global 'correction factor'. Benchmark code is included and can be run at any time to adapt costs to low-level implementation changes. - reimplements flowcontrol.ClientManager in a cleaner and more efficient way, with added capabilities: There is now control over bandwidth, which allows using the flow control parameters for client prioritization. Target utilization over 100 percent is now supported to model concurrent request processing. Total serving bandwidth is reduced during block processing to prevent database contention. - implements an RPC API for the LES servers allowing server operators to assign priority bandwidth to certain clients and change prioritized status even while the client is connected. The new API is meant for cases where server operators charge for LES using an off-protocol mechanism. - adds a unit test for the new client manager. - adds an end-to-end test using the network simulator that tests bandwidth control functions through the new API.
2019-02-26 11:32:48 +00:00
p.fcParams = params
p.fcServer = flowcontrol.NewServerNode(params, &mclock.System{})
2016-10-14 03:51:29 +00:00
p.fcCosts = MRC.decode()
}
p.headInfo = &announceData{Td: rTd, Hash: rHash, Number: rNum}
2016-10-14 03:51:29 +00:00
return nil
}
les, les/flowcontrol: improved request serving and flow control (#18230) This change - implements concurrent LES request serving even for a single peer. - replaces the request cost estimation method with a cost table based on benchmarks which gives much more consistent results. Until now the allowed number of light peers was just a guess which probably contributed a lot to the fluctuating quality of available service. Everything related to request cost is implemented in a single object, the 'cost tracker'. It uses a fixed cost table with a global 'correction factor'. Benchmark code is included and can be run at any time to adapt costs to low-level implementation changes. - reimplements flowcontrol.ClientManager in a cleaner and more efficient way, with added capabilities: There is now control over bandwidth, which allows using the flow control parameters for client prioritization. Target utilization over 100 percent is now supported to model concurrent request processing. Total serving bandwidth is reduced during block processing to prevent database contention. - implements an RPC API for the LES servers allowing server operators to assign priority bandwidth to certain clients and change prioritized status even while the client is connected. The new API is meant for cases where server operators charge for LES using an off-protocol mechanism. - adds a unit test for the new client manager. - adds an end-to-end test using the network simulator that tests bandwidth control functions through the new API.
2019-02-26 11:32:48 +00:00
// updateFlowControl updates the flow control parameters belonging to the server
// node if the announced key/value set contains relevant fields
func (p *peer) updateFlowControl(update keyValueMap) {
if p.fcServer == nil {
return
}
params := p.fcParams
updateParams := false
if update.get("flowControl/BL", &params.BufLimit) == nil {
updateParams = true
}
if update.get("flowControl/MRR", &params.MinRecharge) == nil {
updateParams = true
}
if updateParams {
p.fcParams = params
p.fcServer.UpdateParams(params)
}
var MRC RequestCostList
if update.get("flowControl/MRC", &MRC) == nil {
p.fcCosts = MRC.decode()
}
}
2016-10-14 03:51:29 +00:00
// String implements fmt.Stringer.
func (p *peer) String() string {
return fmt.Sprintf("Peer %s [%s]", p.id,
fmt.Sprintf("les/%d", p.version),
)
}
// peerSetNotify is a callback interface to notify services about added or
// removed peers
type peerSetNotify interface {
registerPeer(*peer)
unregisterPeer(*peer)
}
2016-10-14 03:51:29 +00:00
// peerSet represents the collection of active peers currently participating in
// the Light Ethereum sub-protocol.
type peerSet struct {
peers map[string]*peer
lock sync.RWMutex
notifyList []peerSetNotify
closed bool
2016-10-14 03:51:29 +00:00
}
// newPeerSet creates a new peer set to track the active participants.
func newPeerSet() *peerSet {
return &peerSet{
peers: make(map[string]*peer),
}
}
// notify adds a service to be notified about added or removed peers
func (ps *peerSet) notify(n peerSetNotify) {
ps.lock.Lock()
ps.notifyList = append(ps.notifyList, n)
peers := make([]*peer, 0, len(ps.peers))
for _, p := range ps.peers {
peers = append(peers, p)
}
ps.lock.Unlock()
for _, p := range peers {
n.registerPeer(p)
}
}
2016-10-14 03:51:29 +00:00
// Register injects a new peer into the working set, or returns an error if the
// peer is already known.
func (ps *peerSet) Register(p *peer) error {
ps.lock.Lock()
if ps.closed {
ps.lock.Unlock()
2016-10-14 03:51:29 +00:00
return errClosed
}
if _, ok := ps.peers[p.id]; ok {
ps.lock.Unlock()
2016-10-14 03:51:29 +00:00
return errAlreadyRegistered
}
ps.peers[p.id] = p
p.sendQueue = newExecQueue(100)
peers := make([]peerSetNotify, len(ps.notifyList))
copy(peers, ps.notifyList)
ps.lock.Unlock()
for _, n := range peers {
n.registerPeer(p)
}
2016-10-14 03:51:29 +00:00
return nil
}
// Unregister removes a remote peer from the active set, disabling any further
// actions to/from that particular entity. It also initiates disconnection at the networking layer.
2016-10-14 03:51:29 +00:00
func (ps *peerSet) Unregister(id string) error {
ps.lock.Lock()
if p, ok := ps.peers[id]; !ok {
ps.lock.Unlock()
2016-10-14 03:51:29 +00:00
return errNotRegistered
} else {
delete(ps.peers, id)
peers := make([]peerSetNotify, len(ps.notifyList))
copy(peers, ps.notifyList)
ps.lock.Unlock()
for _, n := range peers {
n.unregisterPeer(p)
}
p.sendQueue.quit()
p.Peer.Disconnect(p2p.DiscUselessPeer)
return nil
2016-10-14 03:51:29 +00:00
}
}
// AllPeerIDs returns a list of all registered peer IDs
func (ps *peerSet) AllPeerIDs() []string {
ps.lock.RLock()
defer ps.lock.RUnlock()
res := make([]string, len(ps.peers))
idx := 0
2017-01-06 14:52:03 +00:00
for id := range ps.peers {
2016-10-14 03:51:29 +00:00
res[idx] = id
idx++
}
return res
}
// Peer retrieves the registered peer with the given id.
func (ps *peerSet) Peer(id string) *peer {
ps.lock.RLock()
defer ps.lock.RUnlock()
return ps.peers[id]
}
// Len returns if the current number of peers in the set.
func (ps *peerSet) Len() int {
ps.lock.RLock()
defer ps.lock.RUnlock()
return len(ps.peers)
}
// BestPeer retrieves the known peer with the currently highest total difficulty.
func (ps *peerSet) BestPeer() *peer {
ps.lock.RLock()
defer ps.lock.RUnlock()
var (
bestPeer *peer
bestTd *big.Int
)
for _, p := range ps.peers {
if td := p.Td(); bestPeer == nil || td.Cmp(bestTd) > 0 {
bestPeer, bestTd = p, td
}
}
return bestPeer
}
// AllPeers returns all peers in a list
func (ps *peerSet) AllPeers() []*peer {
ps.lock.RLock()
defer ps.lock.RUnlock()
list := make([]*peer, len(ps.peers))
i := 0
for _, peer := range ps.peers {
list[i] = peer
i++
}
return list
}
// Close disconnects all peers.
// No new peers can be registered after Close has returned.
func (ps *peerSet) Close() {
ps.lock.Lock()
defer ps.lock.Unlock()
for _, p := range ps.peers {
p.Disconnect(p2p.DiscQuitting)
}
ps.closed = true
}